Clustering of imbalanced high-dimensional media data

Media content in large repositories usually exhibits multiple groups of strongly varying sizes. Media of potential interest often form notably smaller groups. Such media groups differ so much from the remaining data that it may be worthy to look at them in more detail. In contrast, media with popular content appear in larger groups. Identifying groups of varying sizes is addressed by clustering of imbalanced data. Clustering highly imbalanced media groups is additionally challenged by the high dimensionality of the underlying features. In this paper, we present the imbalanced clustering (IClust) algorithm designed to reveal group structures in high-dimensional media data. IClust employs an existing clustering method in order to find an initial set of a large number of potentially highly pure clusters which are then successively merged. The main advantage of IClust is that the number of clusters does not have to be pre-specified and that no specific assumptions about the cluster or data characteristics need to be made. Experiments on real-world media data demonstrate that in comparison to existing methods, IClust is able to better identify media groups, especially groups of small sizes.     

A survey of berth allocation and quay crane scheduling problems in container terminals

Due to the variety of technical equipments and terminal layouts, research has produced a multitude of optimization models for seaside operations planning in container terminals. To provide a support in modeling problem characteristics and in suggesting applicable algorithms this paper reviews the relevant literature. For this purpose new classification schemes for berth allocation problems and quay crane scheduling problems are developed. Particular focus is put on integrated solution approaches which receive increasing importance for the terminal management.     

On finite unions of ideals and cosets

Abstract

Although it is well known that the growth of Thompson's group F is exponential, the exact growth function is still unknown. Elements of its submonoid of positive words can be described using a binary rooted tree, whose norm can be computed assigning weights to each caret. Combining this fact with a combinatorial argument, the growth function of the submonoid is computed and thus providing a first step in the computation of the growth function of the group, as well as a lower bound for the growth rate for the group.     

Multi-objective decision analysis for competence-oriented project portfolio selection

This paper develops a multi-objective optimization model for project portfolio selection taking employee competencies and their evolution into account. The objectives can include economic gains as well as gains expressed in terms of aggregated competence increments according to pre-defined profiles. In order to determine Pareto-optimal solutions, the overall problem is decomposed into a master problem addressing the portfolio selection itself, and a slave problem dealing with a suitable assignment of personnel to the work packages of the selected projects over time. We provide an asymptotic approximation of the problem by a linearized formulation, which allows an efficient and exact solution of the slave problem. For the solution of the master problem, we compare the multi-objective metaheuristics NSGA-II and P-ACO. Experimental results both for synthetically generated test instances and for real-world test instances, based on an application case from the E-Commerce Competence Center Austria, are presented.